Merge pull request #99 from davidgiven/extensions

Validate image extensions before reading, not after.

.appveyor.yml

@@ -15,7 +15,7 @@ install:
 
 build_script:
   - make
-  - zip -9 fluxengine.zip fluxengine.exe brother120tool.exe
+  - zip -9 fluxengine.zip fluxengine.exe brother120tool.exe FluxEngine.cydsn/CortexM3/ARM_GCC_541/Release/FluxEngine.hex
 
 artifacts:
   - path: fluxengine.zip

FluxEngine.cydsn/FluxEngine.cyprj

@@ -3700,6 +3700,6 @@
 </ignored_deps>
 </CyGuid_495451fe-d201-4d01-b22d-5d3f5609ac37>
 <boot_component v="" />
-<current_generation v="60" />
+<current_generation v="68" />
 </CyGuid_fec8f9e8-2365-4bdb-96d3-a4380222e01b>
 </CyXmlSerializer>

doc/building.md

@@ -103,7 +103,43 @@ the unconnected pins and solder a short piece of wire to a GND pin on the
 board. Alternatively you'll need to splice it into your drive's power supply
 cable somehow. (The black one.)
 
-## Building the firmware
+## Programming the board
+
+You've got two options here. You can either use the precompiled firmware
+supplied with the source, or else install the Cypress SDK and build it
+yourself. If you want to hack the firmware source you need the latter, but
+if you trust me to do it for you use the precompiled firmware. In either
+case you'll need Windows and have to install some Cypress stuff.
+
+**Before you read this:** If you're on Windows, good news! You can download a
+precompiled version of the FluxEngine client and precompiled firmware [from
+the GitHub releases
+page](https://github.com/davidgiven/fluxengine/releases/latest). Simply unzip
+it somewhere and run the `.exe` files from a `cmd` window (or other shell).
+Follow the instructions below to program the board with the firmware.
+
+### Using the precompiled firmware
+
+On your Windows machine, [install the PSoC
+Programmer](https://www.cypress.com/products/psoc-programming-solutions).
+**Note:** _not_ the Cypress Programmer, which is for a different board!
+Cypress will make you register.
+
+Once done, run it. Plug the blunt end of the FluxEngine board into a USB
+port (the end which is a USB connector). The programmer should detect it
+and report it as a KitProg. You may be prompted to upgrade the programmer
+hardware; if so, follow the instructions and do it.
+
+Now go to File -> File Load and open
+`FluxEngine.cydsn/CortexM3/ARM_GCC_541/Release/FluxEngine.hex` in the
+project. If you're on Windows, the precompiled zipfile also contains a copy
+of this file. Press the Program button (the one in the toolbar marked with a
+down arrow). Stuff will happen and you should be left with three green boxes
+in the status bar and 'Programming Succeeded' at the top of the log window.
+
+You're done. You can unplug the board and close the programmer.
+
+### Building the firmware yourself
 
 On your Windows machine, [install the Cypress SDK and CY8CKIT-059
 BSP](http://www.cypress.com/documentation/development-kitsboards/cy8ckit-059-psoc-5lp-prototyping-kit-onboard-programmer-and).
@@ -118,7 +154,7 @@ tutorial and making the LED on your board flash. It'll tell you where all the
 controls are and how to program the board. Remember that the big end of the
 board plugs into your computer for programming.
 
-When you're ready, open the `FluxEngine.cydsn/FluxEngine.cywrk` workspace,
+When you're ready, open the `FluxEngine.cydsn/FluxEngine.cyprj` project,
 pick 'Program' from the menu, and the firmware should compile and be
 programmed onto your board.
 
@@ -139,11 +175,6 @@ the port and proceed normally.
 
 ## Building the client
 
-**Before you read this:** If you're on Windows, good news! You can download a
-*precompiled version of the FluxEngine client [from the GitHub releases
-*page](https://github.com/davidgiven/fluxengine/releases/latest). Simply unzip
-*it somewhere and run it from a `cmd` window (or other shell).
-
 The client software is where the intelligence, such as it is, is. It's pretty
 generic libusb stuff and should build and run on Windows, Linux and OSX as
 well, although on Windows it'll need MSYS2 and mingw32. You'll need to

doc/disk-c64.md

@@ -23,7 +23,7 @@ computer](https://ilesj.wordpress.com/2014/05/14/1541-why-so-complicated/) of
 300 bytes per second (!). (The drive itself could transfer data reasonably
 quickly.)
 
-A standard 1541 disk has 35 tracks of 17 to 20 sectors, each 256 bytes long.
+A standard 1541 disk has 35 tracks of 17 to 21 sectors, each 256 bytes long.
 
 Reading discs
 -------------
@@ -34,15 +34,14 @@ Just do:
 fluxengine read c64
 ```
 
-You should end up with an `c64.img` which is 187136 bytes long (for a normal
-1541 disk).
+You should end up with an `c64.d64` file which is 174848 bytes long. You can
+load this straight into a Commodore 64 emulator such as
+[VICE](http://vice-emu.sourceforge.net/).
 
-**Big warning!** The image may not work in an emulator. Commodore 64 disk images are
+**Big warning!** Commodore 64 disk images are
 complicated due to the way the tracks are different sizes and the odd sector
-size. FluxEngine chooses to store them in a simple 256 x 20 x 35 layout,
-with holes where missing sectors should be. This was easiest. If anyone can
-suggest a better way, please [get in
-touch](https://github.com/davidgiven/fluxengine/issues/new).
+size, so you need the special D64 or LDBS output formats to represent them
+sensibly. Don't use IMG unless you know what you're doing.
 
 Useful references
 -----------------

doc/using.md

@@ -148,6 +148,15 @@ based on the extension:
     it doesn't store a lot of the more esoteric LDBS features like format
     types, timings, and data rates.
 
+  - `.d64`: the venerable Commodore 64 disk image format as used by the 1540,
+    1541, etc. This is a special-purpose format due to the weird layout of
+    1540 disks and while you can use this for non-Commodore disks the result
+    will be gibberish. Use this to image Commodore 64 disks and load the
+    result into an emulator.
+
+    FluxEngine's D64 support is currently limited to write only. It will work
+    with up to 40 logical tracks.
+
 ### High density disks
 
 High density disks use a different magnetic medium to low and double density

lib/globals.h

@@ -22,7 +22,7 @@ extern void hexdumpForSrp16(std::ostream& stream, const Bytes& bytes);
 class Error
 {
 public:
-    ~Error()
+    [[ noreturn ]] ~Error()
     {
         std::cerr << "Error: " << _stream.str() << std::endl;
         exit(1);

lib/imagereader/imagereader.cc

@@ -7,6 +7,13 @@
 #include "imagereader/imagereader.h"
 #include "fmt/format.h"
 
+std::map<std::string, ImageReader::Constructor> ImageReader::formats =
+{
+	{".adf", ImageReader::createImgImageReader},
+	{".d81", ImageReader::createImgImageReader},
+	{".img", ImageReader::createImgImageReader},
+};
+
 static bool ends_with(const std::string& value, const std::string& ending)
 {
     if (ending.size() > value.size())
@@ -14,15 +21,29 @@ static bool ends_with(const std::string& value, const std::string& ending)
     return std::equal(ending.rbegin(), ending.rend(), value.rbegin());
 }
 
-std::unique_ptr<ImageReader> ImageReader::create(const ImageSpec& spec)
+ImageReader::Constructor ImageReader::findConstructor(const ImageSpec& spec)
 {
     const auto& filename = spec.filename;
 
-    if (ends_with(filename, ".img") || ends_with(filename, ".adf"))
-        return createImgImageReader(spec);
+	for (const auto& e : formats)
+	{
+		if (ends_with(filename, e.first))
+			return e.second;
+	}
 
-    Error() << "unrecognised image filename extension";
-    return std::unique_ptr<ImageReader>();
+	return NULL;
+}
+
+std::unique_ptr<ImageReader> ImageReader::create(const ImageSpec& spec)
+{
+    verifyImageSpec(spec);
+    return findConstructor(spec)(spec);
+}
+
+void ImageReader::verifyImageSpec(const ImageSpec& spec)
+{
+    if (!findConstructor(spec))
+        Error() << "unrecognised image filename extension";
 }
 
 ImageReader::ImageReader(const ImageSpec& spec):

lib/imagereader/imagereader.h

@@ -12,10 +12,21 @@ public:
 
 public:
     static std::unique_ptr<ImageReader> create(const ImageSpec& spec);
+	static void verifyImageSpec(const ImageSpec& spec);
 
 private:
+	typedef 
+		std::function<
+			std::unique_ptr<ImageReader>(const ImageSpec& spec)
+		>
+		Constructor;
+
+	static std::map<std::string, Constructor> formats;
+
     static std::unique_ptr<ImageReader> createImgImageReader(const ImageSpec& spec);
 
+	static Constructor findConstructor(const ImageSpec& spec);
+
 public:
 	virtual SectorSet readImage() = 0;
 

lib/imagewriter/d64imagewriter.cc

@@ -0,0 +1,64 @@
+#include "globals.h"
+#include "image.h"
+#include "flags.h"
+#include "dataspec.h"
+#include "sector.h"
+#include "sectorset.h"
+#include "imagewriter/imagewriter.h"
+#include "fmt/format.h"
+#include "ldbs.h"
+#include <algorithm>
+#include <iostream>
+#include <fstream>
+
+static int sectors_per_track(int track)
+{
+    if (track < 17)
+        return 21;
+    if (track < 24)
+        return 19;
+    if (track < 30)
+        return 18;
+    return 17;
+}
+
+class D64ImageWriter : public ImageWriter
+{
+public:
+	D64ImageWriter(const SectorSet& sectors, const ImageSpec& spec):
+		ImageWriter(sectors, spec)
+	{}
+
+	void writeImage()
+	{
+		std::cout << "writing D64 triangular image\n";
+
+		std::ofstream outputFile(spec.filename, std::ios::out | std::ios::binary);
+		if (!outputFile.is_open())
+			Error() << "cannot open output file";
+
+        uint32_t offset = 0;
+		for (int track = 0; track < 40; track++)
+		{
+            int sectorCount = sectors_per_track(track);
+            for (int sectorId = 0; sectorId < sectorCount; sectorId++)
+            {
+                const auto& sector = sectors.get(track, 0, sectorId);
+                if (sector)
+                {
+                    outputFile.seekp(offset);
+                    outputFile.write((const char*) sector->data.cbegin(), 256);
+                }
+
+                offset += 256;
+            }
+		}
+    }
+};
+
+std::unique_ptr<ImageWriter> ImageWriter::createD64ImageWriter(
+	const SectorSet& sectors, const ImageSpec& spec)
+{
+    return std::unique_ptr<ImageWriter>(new D64ImageWriter(sectors, spec));
+}
+

lib/imagewriter/imagewriter.cc

@@ -7,6 +7,15 @@
 #include "imagewriter/imagewriter.h"
 #include "fmt/format.h"
 
+std::map<std::string, ImageWriter::Constructor> ImageWriter::formats =
+{
+	{".adf", ImageWriter::createImgImageWriter},
+	{".d64", ImageWriter::createD64ImageWriter},
+	{".d81", ImageWriter::createImgImageWriter},
+	{".img", ImageWriter::createImgImageWriter},
+	{".ldbs", ImageWriter::createLDBSImageWriter},
+};
+
 static bool ends_with(const std::string& value, const std::string& ending)
 {
     if (ending.size() > value.size())
@@ -14,17 +23,29 @@ static bool ends_with(const std::string& value, const std::string& ending)
     return std::equal(ending.rbegin(), ending.rend(), value.rbegin());
 }
 
-std::unique_ptr<ImageWriter> ImageWriter::create(const SectorSet& sectors, const ImageSpec& spec)
+ImageWriter::Constructor ImageWriter::findConstructor(const ImageSpec& spec)
 {
     const auto& filename = spec.filename;
 
-    if (ends_with(filename, ".img") || ends_with(filename, ".adf"))
-        return createImgImageWriter(sectors, spec);
-	else if (ends_with(filename, ".ldbs"))
-		return createLDBSImageWriter(sectors, spec);
+	for (const auto& e : formats)
+	{
+		if (ends_with(filename, e.first))
+			return e.second;
+	}
 
-    Error() << "unrecognised image filename extension";
-    return std::unique_ptr<ImageWriter>();
+	return NULL;
+}
+
+std::unique_ptr<ImageWriter> ImageWriter::create(const SectorSet& sectors, const ImageSpec& spec)
+{
+	verifyImageSpec(spec);
+	return findConstructor(spec)(sectors, spec);
+}
+
+void ImageWriter::verifyImageSpec(const ImageSpec& spec)
+{
+	if (!findConstructor(spec))
+		Error() << "unrecognised image filename extension";
 }
 
 ImageWriter::ImageWriter(const SectorSet& sectors, const ImageSpec& spec):

lib/imagewriter/imagewriter.h

@@ -12,12 +12,25 @@ public:
 
 public:
     static std::unique_ptr<ImageWriter> create(const SectorSet& sectors, const ImageSpec& spec);
+	static void verifyImageSpec(const ImageSpec& filename);
 
 private:
+	typedef 
+		std::function<
+			std::unique_ptr<ImageWriter>(const SectorSet& sectors, const ImageSpec& spec)
+		>
+		Constructor;
+
+	static std::map<std::string, Constructor> formats;
+
     static std::unique_ptr<ImageWriter> createImgImageWriter(
 		const SectorSet& sectors, const ImageSpec& spec);
     static std::unique_ptr<ImageWriter> createLDBSImageWriter(
 		const SectorSet& sectors, const ImageSpec& spec);
+    static std::unique_ptr<ImageWriter> createD64ImageWriter(
+		const SectorSet& sectors, const ImageSpec& spec);
+
+	static Constructor findConstructor(const ImageSpec& spec);
 
 public:
 	virtual void adjustGeometry();

lib/reader.cc

@@ -14,6 +14,7 @@
 #include "bytes.h"
 #include "decoders/rawbits.h"
 #include "track.h"
+#include "imagewriter/imagewriter.h"
 #include "fmt/format.h"
 
 FlagGroup readerFlags { &hardwareFluxSourceFlags, &fluxmapReaderFlags };
@@ -153,7 +154,8 @@ static void replace_sector(std::unique_ptr<Sector>& replacing, Sector& replaceme
 void readDiskCommand(AbstractDecoder& decoder)
 {
 	const ImageSpec outputSpec(output);
-
+	ImageWriter::verifyImageSpec(outputSpec);
+        
 	bool failures = false;
 	SectorSet allSectors;
 	auto tracks = readTracks();

mkninja.sh

@@ -139,6 +139,7 @@ buildlibrary libfmt.a \
 buildlibrary libbackend.a \
 	lib/imagereader/imagereader.cc \
 	lib/imagereader/imgimagereader.cc \
+	lib/imagewriter/d64imagewriter.cc \
 	lib/imagewriter/imagewriter.cc \
 	lib/imagewriter/imgimagewriter.cc \
 	lib/imagewriter/ldbsimagewriter.cc \

src/fe-readc64.cc

@@ -16,7 +16,7 @@ static FlagGroup flags { &readerFlags };
 int mainReadC64(int argc, const char* argv[])
 {
 	setReaderDefaultSource(":t=0-79x2:s=0");
-	setReaderDefaultOutput("c64.img");
+	setReaderDefaultOutput("c64.d64");
     setReaderRevolutions(2);
     flags.parseFlags(argc, argv);